From: Riku Voipio <riku.voipio@iki.fi>
To: Peter Maydell <peter.maydell@linaro.org>
Cc: qemu-devel@nongnu.org, patches@linaro.org
Subject: Re: [Qemu-devel] [PATCH] linux-user: Implement capget, capset
Date: Mon, 17 Mar 2014 13:51:13 +0200 [thread overview]
Message-ID: <20140317115113.GA12134@afflict.kos.to> (raw)
In-Reply-To: <1394820650-14750-1-git-send-email-peter.maydell@linaro.org>
On Fri, Mar 14, 2014 at 06:10:50PM +0000, Peter Maydell wrote:
> Implement the capget and capset syscalls. This is useful because
> simple programs like 'ls' try to use it in AArch64
I'm not seing this with ubuntu trusty, wookeys debian or my
static busybox. Where is your ls from? Also, runnning qemu-linux
user as root? How very brave :)
> , and otherwise
> we emit a lot of noise about it being unimplemented.
Well, it seems gcc 4.8 isn't smart enough for this patch:
linux-user/syscall.c: In function ‘do_syscall’:
linux-user/syscall.c:7739:46: error: ‘target_data’ may be used
uninitialized in this function [-Werror=maybe-uninitialized]
target_data[i].effective = tswap32(data[i].effective);
^
> Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
> ---
> Bugfix or feature? You decide :-)
perhaps unimplemented_nowarn for now and a proper implementation for 2.1 ?
Riku
> linux-user/syscall.c | 71 +++++++++++++++++++++++++++++++++++++++++++++--
> linux-user/syscall_defs.h | 11 ++++++++
> 2 files changed, 80 insertions(+), 2 deletions(-)
>
> diff --git a/linux-user/syscall.c b/linux-user/syscall.c
> index 2a8b66c..53c3d69 100644
> --- a/linux-user/syscall.c
> +++ b/linux-user/syscall.c
> @@ -43,6 +43,7 @@
> #include <sys/resource.h>
> #include <sys/mman.h>
> #include <sys/swap.h>
> +#include <sys/capability.h>
This is from libcap-dev, which might not be installed by default. The
actual capset/capget functions seems to be in libc.
> #include <signal.h>
> #include <sched.h>
> #ifdef __ia64__
> @@ -7641,9 +7642,75 @@ abi_long do_syscall(void *cpu_env, int num, abi_long arg1,
> unlock_user(p, arg1, ret);
> break;
> case TARGET_NR_capget:
> - goto unimplemented;
> case TARGET_NR_capset:
> - goto unimplemented;
> + {
> + struct target_user_cap_header *target_header;
> + struct target_user_cap_data *target_data;
> + struct __user_cap_header_struct header;
> + struct __user_cap_data_struct data[2];
> + struct __user_cap_data_struct *dataptr = NULL;
> + int i, target_datalen;
> + int data_items = 1;
> +
> + if (!lock_user_struct(VERIFY_WRITE, target_header, arg1, 1)) {
> + goto efault;
> + }
> + header.version = tswap32(target_header->version);
> + header.pid = tswap32(target_header->pid);
> +
> + if (header.version != _LINUX_CAPABILITY_VERSION_1) {
> + /* Version 2 and up takes pointer to two user_data structs */
> + data_items = 2;
> + }
> +
> + target_datalen = sizeof(*target_data) * data_items;
> +
> + if (arg2) {
> + if (num == TARGET_NR_capget) {
> + target_data = lock_user(VERIFY_WRITE, arg2, target_datalen, 0);
> + } else {
> + target_data = lock_user(VERIFY_READ, arg2, target_datalen, 1);
> + }
> + if (!target_data) {
> + unlock_user_struct(target_header, arg1, 0);
> + goto efault;
> + }
> +
> + if (num == TARGET_NR_capset) {
> + for (i = 0; i < data_items; i++) {
> + data[i].effective = tswap32(target_data[i].effective);
> + data[i].permitted = tswap32(target_data[i].permitted);
> + data[i].inheritable = tswap32(target_data[i].inheritable);
> + }
> + }
> +
> + dataptr = data;
> + }
> +
> + if (num == TARGET_NR_capget) {
> + ret = get_errno(capget(&header, dataptr));
> + } else {
> + ret = get_errno(capset(&header, dataptr));
> + }
> +
> + /* The kernel always updates version for both capget and capset */
> + target_header->version = tswap32(header.version);
> + unlock_user_struct(target_header, arg1, 1);
> +
> + if (arg2) {
> + if (num == TARGET_NR_capget) {
> + for (i = 0; i < data_items; i++) {
> + target_data[i].effective = tswap32(data[i].effective);
> + target_data[i].permitted = tswap32(data[i].permitted);
> + target_data[i].inheritable = tswap32(data[i].inheritable);
> + }
> + unlock_user(target_data, arg2, target_datalen);
> + } else {
> + unlock_user(target_data, arg2, 0);
> + }
> + }
> + break;
> + }
> case TARGET_NR_sigaltstack:
> #if defined(TARGET_I386) || defined(TARGET_ARM) || defined(TARGET_MIPS) || \
> defined(TARGET_SPARC) || defined(TARGET_PPC) || defined(TARGET_ALPHA) || \
> diff --git a/linux-user/syscall_defs.h b/linux-user/syscall_defs.h
> index 732c9e3..7db878a 100644
> --- a/linux-user/syscall_defs.h
> +++ b/linux-user/syscall_defs.h
> @@ -2559,3 +2559,14 @@ struct target_sigevent {
> } _sigev_thread;
> } _sigev_un;
> };
> +
> +struct target_user_cap_header {
> + uint32_t version;
> + int pid;
> +};
> +
> +struct target_user_cap_data {
> + uint32_t effective;
> + uint32_t permitted;
> + uint32_t inheritable;
> +};
> --
> 1.9.0
next prev parent reply other threads:[~2014-03-17 11:51 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-03-14 18:10 [Qemu-devel] [PATCH] linux-user: Implement capget, capset Peter Maydell
2014-03-17 11:51 ` Riku Voipio [this message]
2014-03-17 12:12 ` Peter Maydell
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20140317115113.GA12134@afflict.kos.to \
--to=riku.voipio@iki.fi \
--cc=patches@linaro.org \
--cc=peter.maydell@linaro.org \
--cc=qemu-devel@nongnu.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).